Validation of normalized singular spectrum area as a classifier for targeted microbubble enhancement

Molecularly targeted microbubbles (MBs), comprising a gas core and a functionalized shell, 1-5 μm in diameter, enable visualization of disease marker concentration using a method termed ultrasound molecular imaging (USMI). The current state-of-the-art method for USMI based quantification of molecular markers is differential targeted enhancement (dTE), which employs destructive pulses of ultrasound to quantify the amount of targeted MB adherence. In this study, we sought to quantify the signal from adherent MBs non-destructively using a statistical parameter termed normalized singular spectrum area (NSSA) in a murine tumor model. The sensitivity and specificity of NSSA-based signal classification was compared to matched dTE measurements in a mouse hindlimb tumor.